DNA in eukaryotic cells is tightly complexed with proteins (histones) to form chromatin. Histones are small, positively charged proteins which are rich in basic amino acids (positively charged at physiological pH), which contact the phosphate groups (negatively charged at physiological pH) of DNA. There are five main classes of histones, H1, H2A, H2B, H3, and H4. The amino acid sequences of histones H2A, H2B, H3, and H4 show remarkable conservation between species, whereas H1 varies somewhat, and in some cases is replaced by another histone, e.g., H5. Four pairs of each of H2A, H2B, H3, and H4 together form a disk-shaped octomeric protein core, around which DNA (about 140 base pairs) is wound to form a nucleosome. Individual nucleosomes are connected by short stretches of linker DNA associated with another histone molecule (e.g., H1, or in certain cases, H5) to form a structure resembling a beaded string, which is itself arranged in a helical stack, known as a solenoid.
The majority of histones are synthesised during the S phase of the cell cycle, and newly synthesised histones quickly enter the nucleus to become associated with DNA. Within minutes of its synthesis, new DNA becomes associated with histones in nucleosomal structures.
A small fraction of histones, more specifically, the amino side chains thereof, are enzymatically modified by post-translational addition of methyl, acetyl, or phosphate groups, neutralising the positive charge of the side chain, or converting it to a negative charge. For example, lysine and arginine groups may be methylated, lysine groups may be acetylated, and serine groups may be phosphorylated. For lysine, the —(CH2)4—NH2 sidechain may be acetylated, for example by an acetyltransferase enzyme, to give the amide —(CH2)4—NHC(═O)CH3. Methylation, acetylation, and phosphorylation of amino termini of histones which extend from the nucleosomal core affects chromatin structure and gene expression. (See, for example, Spencer and Davie, 1999).
Acetylation and deacetylation of histones is associated with transcriptional events leading to cell proliferation and/or differentiation. Regulation of the function of transcription factors is also mediated through acetylation. Recent reviews of histone deacetylation include Kouzarides, 1999 and Pazin et al., 1997.
The correlation between the acetylation status of histones and the transcription of genes has been known for over 30 years (see, for example, Howe et al., 1999). Certain enzymes, specifically acetylases (e.g., histone acetyltransferase, HAT) and deacetylases (e.g., histone deacetylase, HDAC), which regulate the acetylation state of histones have been identified in many organisms and have been implicated in the regulation of numerous genes, confirming the link between acetylation and transcription. See, for example, Davie, 1998. In general, histone acetylation correlates with transcriptional activation, whereas histone deacetylation is associated with gene repression.
A growing number of histone deacetylases (HDACs) have been identified (see, for example, Ng and Bird, 2000). The first deacetylase, HDAC1, was identified in 1996 (see, for example, Tauton et al., 1996). Subsequently, two other nuclear mammalian deacetylases has been found, HDAC2 and HDAC3 (see, for example, Yang et al., 1996, 1997, and Emiliani et al., 1998). See also, Grozinger et al., 1999; Kao et al., 2000; and Van den Wyngaert et al., 2000.
Eight human HDACs have been cloned so far:                HDAC1 (Genbank Accession No. NP—004955)        HDAC2 (Genbank Accession No. NP—001518)        HDAC3 (Genbank Accession No. O15739)        HDAC4 (Genbank Accession No. AAD29046)        HDAC5 (Genbank Accession No. NP—005465)        HDAC6 (Genbank Accession No. NP—006035)        HDAC7 (Genbank Accession No. AAF63491)        HDAC8 (Genbank Accession No. AAF73428)        
These eight human HDACs fall in two distinct classes: HDACs 1,2,3 and 8 are in class I, and HDACs 4,5,6 and 7 are in class II.
There are a number of histone deacetylases in yeast, including the following:                RPD3 (Genbank Accession No. NP—014069)        HDA1 (Genbank Accession No. P53973)        HOS1 (Genbank Accession No. Q12214)        HOS2 (Genbank Accession No. P53096)        HOS3 (Genbank Accession No. Q02959)        
There are also numerous plant deacetylases, for example, HD2, in Zea mays (Genbank Accession No. AF254073—1).
HDACs function as part of large multiprotein complexes, which are tethered to the promoter and repress transcription. Well characterised transcriptional repressors such as Mad (Laherty et al., 1997), pRb (Brehm et al., 1998), nuclear receptors (Wong et al., 1998) and YY1 (Yang et al., 1997) associate with HDAC complexes to exert their repressor function.
The study of inhibitors of histone deacetylases indicates that these enzymes play an important role in cell proliferation and differentiation. The inhibitor Trichostatin A (TSA) (Yoshida et al., 1990a) causes cell cycle arrest at both G1 and G2 phases (Yoshida and Beppu, 1988), reverts the transformed phenotype of different cell lines, and induces differentiation of Friend leukaemia cells and others (Yoshida et al., 1990b). TSA (and SAHA) have been reported to inhibit cell growth, induce terminal differentiation, and prevent the formation of tumours in mice (Finnin et al., 1999).

Cell cycle arrest by TSA correlates with an increased expression of gelsolin (Hoshikawa et al., 1994), an actin regulatory protein that is down regulated in malignant breast cancer (Mielnicki et al., 1999). Similar effects on cell cycle and differentiation have been observed with a number of deacetylase inhibitors (Kim et al., 1999).
Trichostatin A has also been reported to be useful in the treatment of fibrosis, e.g., liver fibrosis and liver cirrhosis. See, e.g., Geerts et al., 1998.
Recently, certain compounds that induce differentiation have been reported to inhibit histone deacetylases. Several experimental antitumour compounds, such as trichostatin A (TSA), trapoxin, suberoylanilide hydroxamic acid (SAHA), and phenylbutyrate have been reported to act, at least in part, by inhibiting histone deacetylase (see, e.g., Yoshida et al., 1990; Richon et al., 1998; Kijima et al., 1993). Additionally, diallyl sulfide and related molecules (see, e.g., Lea et al., 1999), oxamflatin (see, e.g., Kim et al., 1999), MS-27-275, a synthetic benzamide derivative (see, e.g., Saito et al., 1999; Suzuki et al., 1999; note that MS-27-275 was later re-named as MS-275), butyrate derivatives (see, e.g., Lea and Tulsyan, 1995), FR901228 (see, e.g., Nokajima et al., 1998), depudecin (see, e.g., Kwon et al., 1998), and m-carboxycinnamic acid bishydroxamide (see, e.g., Richon et al., 1998) have been reported to inhibit histone deacetylases. In vitro, some of these compounds are reported to inhibit the growth of fibroblast cells by causing cell cycle arrest in the G1 and G2 phases, and can lead to the terminal differentiation and loss of transforming potential of a variety of transformed cell lines (see, e.g.,
Richon et al, 1996; Kim et al., 1999; Yoshida et al., 1995; Yoshida & Beppu, 1988). In vivo, phenybutyrate is reported to be effective in the treatment of acute promyelocytic leukemia in conjunction with retinoic acid (see, e.g., Warrell et al., 1998). SAHA is reported to be effective in preventing the formation of mammary tumours in rats, and lung tumours in mice (see, e.g., Desai et al., 1999).
The clear involvement of HDACs in the control of cell proliferation and differentiation suggest that aberrant HDAC activity may play a role in cancer. The most direct demonstration that deacetylases contribute to cancer development comes from the analysis of different acute promyelocytic leukaemias (APL). In most APL patients, a translocation of chromosomes 15 and 17 (t(15;17)) results in the expression of a fusion protein containing the N-terminal portion of PML gene product linked to most of RARα (retinoic acid receptor). In some cases, a different translocation (t(11;17)) causes the fusion between the zinc finger protein PLZF and RARα. In the absence of ligand, the wild type RARα represses target genes by tethering HDAC repressor complexes to the promoter DNA. During normal hematopoiesis, retinoic acid (RA) binds RARα and displaces the repressor complex, allowing expression of genes implicated in myeloid differentiation. The RARα fusion proteins occurring in APL patients are no longer responsive to physiological levels of RA and they interfere with the expression of the RA-inducible genes that promote myeloid differentiation. This results in a clonal expansion of promyelocytic cells and development of leukaemia. In vitro experiments have shown that TSA is capable of restoring RA-responsiveness to the fusion RARα proteins and of allowing myeloid differentiation. These results establish a link between HDACs and oncogenesis and suggest that HDACs are potential targets for pharmaceutical intervention in APL patients. (See, for example, Kitamura et al., 2000; David et al., 1998; Lin et al., 1998).
Furthermore, different lines of evidence suggest that HDACs may be important therapeutic targets in other types of cancer. Cell lines derived from many different cancers (prostate, colorectal, breast, neuronal, hepatic) are induced to differentiate by HDAC inhibitors (Yoshida and Horinouchi, 1999). A number of HDAC inhibitors have been studied in animal models of cancer. They reduce tumour growth and prolong the lifespan of mice bearing different types of transplanted tumours, including melanoma, leukaemia, colon, lung and gastric carcinomas, etc. (Ueda et al., 1994; Kim et al., 1999).
Psoriasis is a common chronic disfiguring skin disease which is characterised by well-demarcated, red, hardened scaly plaques: these may be limited or widespread. The prevalence rate of psoriasis is approximately 2%, i.e., 12.5 million sufferers in the triad countries (US/Europe/Japan). While the disease is rarely fatal, it clearly has serious detrimental effects upon the quality of life of the patient: this is further compounded by the lack of effective therapies. Present treatments are either ineffective, cosmetically unacceptable, or possess undesired side effects. There is therefore a large unmet clinical need for effective and safe drugs for this condition.
Psoriasis is a disease of complex etiology. Whilst there is clearly a genetic component, with a number of gene loci being involved, there are also undefined environmental triggers. Whatever the ultimate cause of psoriasis, at the cellular level, it is characterised by local T-cell mediated inflammation, by keratinocyte hyperproliferation, and by localised angiogenesis. These are all processes in which histone deacetylases have been implicated (see, e.g., Saunders et al., 1999; Bernhard et al, 1999; Takahashi et al, 1996; Kim et al, 2001). Therefore HDAC inhibitors may be of use in therapy for psoriasis. Candidate drugs may be screened, for example, using proliferation assays with T-cells and/or keratinocytes.
Thus, one aim of the present invention is the provision of compounds which are potent inhibitors of histone deacetylases (HDACs). There is a pressing need for such compounds, particularly for use as antiproliferatives, for example, anti-cancer agents, agents for the treatment of psoriasis, etc.
Such molecules desirably have one or more of the following properties and/or effects:                (a) easily gain access to and act upon tumour cells;        (b) down-regulate HDAC activity;        (c) inhibit the formation of HDAC complexes;        (d) inhibit the interactions of HDAC complexes;        (e) inhibit tumour cell proliferation;        (e) promote tumour cell apoptosis;        (f) inhibit tumour growth; and,        (g) complement the activity of traditional chemotherapeutic agents.        
A number of carbamic acid compounds have been described.
Amides
Hashimoto et al., 1989 describe hydroxamic acid compounds which are claimed to inhibit cell proliferation. Some of the compounds are carbamic acid compounds having a substituted phenyl-dione group linked to a carbamic acid group (—CONHOH) via an aryl-substituted alkylene group.
Ohtani et al., 1993 describe a number of hydroxamic acid compounds which are claimed to be inhibitors of ras transformation. A few of the compounds are carbamic acid compounds having a phenylacylamido group (—NHCOPh) linked to a carbamic acid group (—CONHOH) via a phenylene-meta-alkylene group having a carbon-carbon triple bond. See, for example, compounds I-29 (page 69), I-39 (page 87), and I-41 (page 90). Compound I-41, shown below, employs an aryl leader.

Onishi et al., 1996, describe several hydroxamic acid compounds which have a phenyl (or substituted phenyl) group linked via an oxazole group to a carbamic acid group. These compounds were reported to inhibit a deacetylase enzyme critical in the biosynthesis of lipid A (a component of the outer membrane of Gram-negative bacteria).
Parsons et al., 1998 describe a number of hydroxamic acid compounds which are claimed to selectively prevent the growth of a variety of human tumour cell lines.
Some of the compounds are carbamic acid compounds having an arylamide group linked to a carbamic acid group via a methylene or substituted methylene group (see, for example, pages 16 and 17).
Some of the compounds are carbamic acid compounds having a phenylamido group (—CONHPh) linked to a carbamic acid group (—CONHOH) via a long alkylene chain, —(CH2)n—, wherein n is from 4 to 7 (see, for example, pages 47, 48, and 58 therein).
Some of the compounds are carbamic acid compounds having an aryl group linked via a short chain to an amide group (—CONH—), which in turn is linked via a short chain (e.g., 3 atoms or less) to a carbamic acid group (—CONHOH). See, for example, page 16, 2nd formula; page 46, 4th formula; page 51, compound 7; and page 61, 2nd formula.

Richon et al., 1998 describe several hydroxamic acid compounds, including SAHA, which apparently inhibit HDAC activity, and induce terminal differentiation and/or apoptosis in various transformed cells (see, for example, Table 1 therein).
Suzuki et al., 1998 describe a number of hydroxamic acid compounds which are claimed to have antitumour activity. Some of the compounds are carbamic acid compounds having a substituted phenylamido group (—CONHPh) linked to a carbamic acid (—CONHOH) group via a phenylene-meta-ethenylene or phenylene-para-ethylene group (see, for example, pages 8 and 9, compounds 31-50).
Breslow et al., 1994, 1995, 1997 describe a number of hydroxamic acid compounds which are claimed to selectively induce terminal differentiation of neoplastic cells.
Some of the compounds are carbamic acid compounds having a substituted phenylacylamido group (—NHCOPh) linked to a carbamic acid (—CONHOH) group via a long alkylene chain, —(CH2)n—, wherein n is from 4 to 8
Some of the compounds are carbamic acid compounds having a substituted phenylamido group (—CONHPh) or phenylacylamido group (—NHCOPh) linked to a carbamic acid (—CONHOH) group via a long alkylene chain, —(CH2)n—, wherein n is from 4 to 8 (see, for example, columns 7 and 13 of Breslow et al., 1997), or via a phenylene group (see, for example, columns 24, 30-31 and compounds 20-55 in Table 1 of Breslow et al., 1997).
One of the compounds is a carbamic acid compound having benzylamido group (—CONHCH2Ph) linked to a carbamic acid group (—CONHOH) via a —(CH2)6— group (see, for example, compound 19 in Table 1, at column 37 of Breslow et al., 1997).
Jung et al., 1997, 1999, describe several aromatic hydroxamic acid compounds which apparently inhibit HDAC. Some of the compounds have a phenylamido group (PhCONH—). One compound, a peptide analog, is shown below (see, e.g., compound 6 in Jung et al., 1997; compound 4 in Jung et al., 1999).

Kato et al., 1998, describe a number of aromatic hydroxamic acid compounds, comprising an aryl group linked via an alkylene group to a carbamic acid group, which are apparently active in the treatment of neurodegenerative conditions. One compound, 4-1 at columns 63-64, has a phenylamido group (PhCONH—) linked via a —(CH2)5— group to a carbamic acid group.
Glick et al., 1999, describe the apparent apoptotic and differentiating effects of m-carboxy-cinnamic acid bishydroxamide (CBHA) on various tumour cell lines.
Massa et al., 2001, describe various hydroxamic acid compounds which have a benzoyl (or substituted benzoyl) group linked via a pyrrolyl group and an C2alkylene group (—CH═CH— or —CH2CH2—) to a carbamic acid group. The compounds apparently showed HDAC inhibitory activity in the micromolar range.
Sulfonamides
Oxamflatin, also known as (2E)-5-[3-[(phenylsulfonyl)amino]phenyl]-pent-2-en-4-ynohydroxamic acid, shown below, has been reported to have in vitro antiproliferative activity against various mouse and human tumour cell lines, and in vivo antitumour activity against B16 melanoma (see, e.g., Sonoda et al., 1996; Kim et al., 1999).

Ohtani et al., 1993, describe a number of hydroxamic acid compounds which are claimed to be inhibitors of ras transformation. Many of the compounds are hydroxmic acid compounds which have a sulfonamide group, and which employ an acid leader which is: a phenylene-ortho-alkylene (e.g., I-10); phenylene-meta-alkylene (e.g., I-24); phenylene-para-alkylene (e.g., I-12); or napthylen-1,2-diyl (e.g., I-20). However, in every case, the sulfonamide group is —SO2NR—, as opposed to —NRSO2—. Also, in every case, the terminal aryl group is linked directly to the —SO2NR— sulfonamide group, without an intervening aryl leader. Ohtani et al., 1996, describe similar compounds.
Richon et al., 2001, describe various branched compounds which apparently inhibit histone deacetylase. See the table at pages 96-101 therein. Some of the compounds are carbamic acid compounds having a carbamic acid group (—CONHOH) linked to a branch point, from which two aryl groups are appended. A few linear carbamic acid compounds are also described, including a single —SO2NH— sulfonamide carbamic acid with a —(CH2)5— acid leader (compound 671).
Delorme et al., 2001, describe various carbamic acid compounds, including compounds having, inter alfa, a sulfonamide group. Of the 108 compounds in the table at pages 114-123 therein, 88 are carbamic acids (—CONHOH), and the remainder are terminal amides, —CONHR. Of the 88 carbamic acid compounds, 54 have a sulfonamide linkage.
Of the 54 sulfonamide carbamic acids, 51 are indicated to have a —SO2NR— sulfonamide group, and 3 (compounds 98, 161, and 162) are indicated to have a —NRSO2— sulfonamide group.
All of the 54 sulfonamide carbamic acids employ a phenylene-alkylene acid leader group (analogous to Q2 herein). Of the 54 compounds, 52 employ a phenylene-para-alkylene group, and only 2 (compounds 41 and 26) employ a phenylene-meta-alkylene group (-Ph-CH2— and -Ph-(CH2)4—, respectively). Compounds 41 and 26 both have a —SO2NR— sulfonamide group, as opposed to a —NRSO2— sulfonamide group; the former has a benzothiophenyl group, and the latter has a phenyl group.
All but one of the 54 sulfonamide carbamic acids have an aryl group linked directly to the sulfonamide; compound 100 has a benzyl group (Ph-CH2—) linked a —SO2NR— sulfonamide group linked to phenylene-para-ethylene.